A Survey of Methods and Architectures for IoT-Based Smart Pharmacies for Optimizing Stock Management with Siamese Heterogeneous Convolutional Neural Networks

Abstract

The integration of Internet of Things (IoT) technologies with artificial intelligence (AI) and deep learning has revolutionized pharmaceutical inventory management by addressing the inefficiencies of traditional, manual stock control systems. Conventional approaches often suffer from inaccurate tracking, frequent stockouts, overstocking, and medication wastage due to expiration. IoT-enabled smart pharmacy systems overcome these limitations through real-time monitoring using sensors, RFID tags, and intelligent storage units, generating continuous data streams for advanced analytics. Deep learning techniques, particularly convolutional neural networks (CNNs) and Siamese Heterogeneous Convolutional Neural Networks (SHCNNs), have demonstrated strong capabilities in analyzing complex pharmaceutical datasets by identifying patterns and similarity relationships between historical and real-time data. These models support accurate demand forecasting, anomaly detection, and optimized stock classification. Furthermore, the integration of IoT with emerging technologies such as blockchain and edge computing enhances supply chain transparency, data security, and decision-making efficiency. This survey emphasizes stock optimization strategies while identifying challenges such as interoperability, implementation costs, and data privacy. Overall, IoT-AI integrated systems significantly improve pharmacy operations, with future research focusing on hybrid models, federated learning, and real-time edge-based solutions.